Rescue of chromosomal T-antigen sequences onto extrachromosomally replicating, defective simian virus 40 DNA by homologous recombination

Targeted homologous recombination in mammalian cells

This article presents a review of recent progress in the field of targeted homologous recombination in mammalian cells. Beginning with an introduction of basic terminology and why 'gene targeting' is potentially such a powerful genetic tool, the article explores some of the obstacles that must be overcome in order for targeting to be generally useful. In particular, the different ways in which investigators have been able to work around the great inefficiency of gene targeting is covered in some detail. When possible, insights into the mechanisms(s) of gene targeting are extracted from the published literature. The use of targeted gene 'knockout' in mouse embryonic stems cells to create animal disease models is discussed. The need for systematic studies into the mechanisms(s) of targeting to make gene targeting useful for human gene therapy is recognized, and some suggestions are made.

Ability of structurally related polycyclic aromatic carcinogens to induce homologous recombination between duplicated chromosomal sequences in mouse L cells

To investigate the role of DNA damage in the induction of homologous recombination in mammalian cells, a series of structurally related, polycyclic aromatic carcinogens, i.e., 1-nitrosopyrene (1-NOP), N-acetoxy-2-acetylaminofluorene (N-AcO-AAF), and 4-nitroquinoline 1-oxide (4-NQO), were compared for their ability to cause intrachromosomal homologous recombination between two herpes simplex virus thymidine kinase (Htk) genes stably integrated in the genome of a tk- mouse L cell strain 333 M. Each Htk gene contains an 8-bp XhoI linker inserted at a unique site so that expression of a functional Htk enzyme requires a productive recombinational event between the two nonfunctional genes. Each carcinogen caused a dose-dependent increase in the frequency of recombination. The results were compared to what had been found previously for a structurally related carcinogen, (+/-)-7 beta, 8 alpha-dihydroxy-9 alpha, 10 alpha-epoxy-7,8,9,10-tetrahydrobenzo[a]pyrene (BPDE). As a function of concentration, BPDE was the most active agent, followed by 4-NQO, and 1-NOP, and then N-AcO-AAF. When compared on the basis of equal cell killing, the most efficient carcinogen was 1-NOP, followed by N-AcO-AAF and BPDE, and then 4-NQO. Use of tritium-labeled compounds to determine the frequency of recombination as a function of the number of adducts initially bound to DNA showed that the most effective agent was BPDE, followed by 1-NOP and 4-NQO, and then N-AcO-AAF (ratio, 6.6:2.5:1.8:1.0). To determine if these differences in recombinagenic effectiveness reflected different rates of removal of the adducts from DNA, we measured the percentage of DNA adducts removed during the 24-h period post treatment and found that 1-NOP, 4-NQO and N-AcO-AAF residues were removed at approximately the same rate, i.e., 25%-30% off. Cellular analysis of a series of independent recombinants indicated that approximately 82% of the recombinational events induced by each agent were consistent with gene conversion. DNA-DNA hybridization analysis confirmed this, and showed that each recombinant tested contained an XhoI-resistant (wild-type) Htk gene; with the majority retaining the Htk gene duplication, consistent with nonreciprocal transfer of wild-type genetic information. In the rest, only a single copy of the Htk gene remained, reflecting a single reciprocal exchange within a chromatid or a single unequal exchange between sister chromatids.

Analysis of recombination in mammalian cells using SV40 and SV40-derived vectors

Viruses and viral vectors have played a crucial role in our understanding of the pathways of homologous and non-homologous recombination in mitotically dividing mammalian cells. In particular, they have allowed the confirmation of the preponderance of non-homologous over homologous recombination events and led to schemes for the selection and isolation of homologous recombination products. These studies have allowed an examination of the properties of reciprocal and non-reciprocal homologous recombination events extrachromosomally, in the chromosome and between plasmids and chromosomes. They suggest that it is feasible now to direct DNA segments to predetermined chromosomal locations by homologous recombination.

Recombination of selectable marker DNA in Nicotiana tabacum

A chimeric neomycin phosphotransferase II (NPT II) gene, which normally provides kanamycin resistance to transformed plant cells, was inactivated by in vitro deletions. Repair plasmids not containing plant-specific transcription signals but containing only the NPT II coding region (or parts of it) were used in co-transformation experiments involving direct DNA uptake into protoplasts isolated from Nicotiana tabacum. Recombination, or gene conversion mediated by homologous sequences produced active NPT II genes in about 1% of transformants, rendering these cells resistant to kanamycin. Analysis of the size of the active enzyme indicated that recombination had occurred producing an NPT II gene indistinguishable from the wild-type gene. Southern blot analysis revealed that the bulk of co-transformed donor plasmid DNA had suffered structural modifications; however, kanamycin resistance was inherited in a Mendelian fashion, indicating that at least one functional and structurally intact copy of the regenerated NPT II gene is integrated into the host genome.

Correction of deletions in mammalian cells by gene conversion

We have constructed substrates to study the conversion of deletions in mammalian cells both extrachromosomally and after the stable integration of the substrates into the chromosome. These substrates were designed to study gene conversion without the complication of reciprocal recombination events. The substrates contain insertion or deletion mutations of the neomycin resistance gene (neo) and an internal, homologous fragment of the neo gene (neo-526), such that gene conversion from neo-526 to the mutated neo gene restores a functional neo gene. We have shown that extrachromosomally insertions of 10 bp or deletions of 22 or 167 bp are converted to wild-type at similar frequencies (1-6 X 10(-4)). Chromosomal gene conversion occurred at frequencies of about 10(-6)-10(-7) per cell generation. As expected from the experimental design, all recombination events analyzed in mammalian cells using these substrates appear to be due to gene conversion.

SV40-based Escherichia coli shuttle vectors infectious for monkey cells

We describe SV40-based Escherichia coli shuttle vectors which can be packaged as pseudovirions without excision of plasmid sequences and which can be rescued in bacteria. These vectors replicate and are transmitted as virus in monkey COS cells without requiring a helper virus. Extrachromosomal vector DNA isolated from infected cells can be rescued in E. coli, so that DNA alterations can be easily screened. Indeed, some of the constructions give rise to very stable plasmids with no detectable rearrangements after multiple lytic cycles in COS cells. The spontaneous mutation frequency measured in bacteria, on the lacO target, is smaller than those usually found with shuttle vectors. We also constructed an expression vector derived from one of our infectious viruses by inserting the chloramphenicol acetyl transferase gene, expressed from the SV40 early promoter, which is efficiently transduced to cells by infection. In this system, the shuttle virus combines the convenience of plasmid rescue and analysis in bacteria, with the advantages of infectious virus.